Concordance of results in evaluating cognitive functions using the Montreal Cognitive Assessment (MoCA) and Picture Naming and Immediate Recall (PICNIR – Hedgehog Version) in elderly institutionalized people

Summary

Aim: The aim of the research was to assess cognitive functions of elderly people in nursing homes using standardized tests –⁠ Montreal Cognitive Assessment (MoCA) and Picture Naming and Immediate Recall (PICNIR). Methodology: The study included 76 clients from three nursing homes. The assessment of cognitive functions was carried out using the standardized tests MoCA and PICNIR (Hedgehog Version). Statistical processing was performed at a significance level of α = 0.05. Results: The average score in MoCA was 20 points, and the average results in PICNIR were 3/5. According to MoCA (with a threshold score of ≤ 24 points), 21% of elderly people had a normal cognitive state, and according to PICNIR, 25% of respondents reached the norm. No statistically significant difference in the level of cognitive functions was found in either the MoCA or PICNIR tests in relation to sex, education, or length of stay in the nursing home. However, a statistically significant difference in the level of cognitive functions was demonstrated in relation to age, cognitive training, and physical exercise, and it was found that individuals undergoing cognitive training and physical exercise achieved better results in both tests. A concordance in the detection of cognitive impairment using MoCA and PICNIR was confirmed in 97% of cases. A significant correlation was found between the MoCA and PICNIR tests. The Pearson correlation coefficient between MoCA and PICNIR –⁠ Mistakes in Naming demonstrated a negative correlation of –0.625 (P < 0.001). The Pearson correlation coefficient between MoCA and PICNIR –⁠ Correctly Recalled Picture Names indicated a positive correlation of 0.86 (P < 0.001). Conclusion: Early detection and monitoring of cognitive deficits using screening tests should be an integral part of providing long-term care for elderly people. The PICNIR test is a suitable choice for early detection of cognitive deficits and can serve as a comparable alternative to the MoCA test.

Keywords:

nursing home – Cognition – assessment – Montreal Cognitive Assessment (MoCA) – Picture Naming and Immediate Recall (PICNIR)

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version.

 

Introduction

Aging results in atrophy of all body structures and aging is accompanied by changes in cognitive performance. Cognitive aging can be successful, normal or pathological. Pathological aging includes mild cognitive impairment and dementia [1]. Early detection of cognitive changes and subsequent intervention can prolong the period of self-sufficiency of the client and reduce the level of assistance and support provided by direct care staff [2,3]. In institutional care, however, it is often the case that the diagnosis of dementia is not adequately made in clients [4].

Currently, there is no uniform methodology or obligation to assess cognitive function in residential services for the elderly [5]. Not only is there no uniform recommendation for cognitive function assessment, but the situation is exacerbated by the fact that there are over 50 screening tests to detect cognitive function [6]. A recommended procedure for the diagnosis of Alzheimer's disease and other dementia-related disorders has been traced in Czech sources. According to this procedure, tests such as the Mini-Mental State Examination (MMSE), the Montreal Cognitive Assessment (MoCA) and the Addenbrooke's Cognitive Examination (ACE) are available, but it is no longer specified which test is most appropriate and who should perform it [7]. International guidelines primarily provide diagnostic procedures for specialist physicians [8,9]. Recommendations from the Ontario Nurses Association provide an overview of assessment tools and their limitations [10] and recommend the use of valid screening tests [10,11].

Based on Prof. Bartoš's recommendation, it is advisable to select one of the very short tests for universal screening that have valid standards for the Czech elderly population. They are freely available and are not time-consuming in the case of universal screening. The innovative Czech tests Picture Naming and Picture Amenities (POBAV) and The Amnesia Light and Brief Assessment (ALBA) seem to be ideal. POBAV is a high quality valid test that has been certified by the Ministry of Health of the Czech Republic since 2017. In case of individuals with visual deficits and greater graphomotor difficulties, the ALBA test should be used. The POBAV test is designed to rapidly assess speech and memory (episodic, semantic, short-term) and detect various cognitive disorders. In the case of borderline scores, it is recommended to test with a more detailed test that includes detection of more types of cognitive functions. The MoCA, which is a good compromise between the MMSE and the ACE, can be used for this purpose. In addition to valid norms for the Czech population (not only the elderly), the MoCA test has the advantage of its high sensitivity for mild cognitive impairment and providing a comprehensive cognitive picture [12,13].

The aim of the research was to assess cognitive function in elderly people in nursing homes through standardized tests MoCA and POBAV.

 

Methodology

Data collection

This was a cross-sectional study and in the context of data collection, a record sheet was created which contained standardised MoCA and POBAV tests (hedgehog version) and other data relating to the research participant (demographic data, educational attainment, length of stay in a home for the elderly, completion of fitness training/week -⁠ individual, group, MOTOmed^® [Jablunka nad Bečvou, Czech Republic] for lower limbs and cognitive training/week -⁠ individual, group).

The MoCA test was freely available on the official website at the time of the research (October 2019 -⁠ February 2020) (from February 2021 its use is restricted by certification) [14]. The cognitive function domains tested are: memory (semantic, verbal, short-term), attention, visual-spatial skills, executive functions, language and speech skills, thinking (abstract), orientation (place, time) [15]. Valid data for the Czech population, including a conversion table of the scales between MoCA and MMSE, was provided by Kopeček et al. [16]. According to the Czech normative study, individuals scoring between 25-30 points on the MoCA -⁠ CZ can be considered healthy, in the range of 24-22 points there is a so-called grey zone indicating a zone of mild cognitive impairment (MCI) or even normal cognition, in the case of 21 points MCI can be determined and a score ≤ 18 detects dementia. When the cutoff score was lowered to ≤ 24, an optimal sensitivity of 87% for AD-MKP and a specificity of 72% were achieved [17]. The original cutoff score based on research in the Czech population was reduced from 26 points to ≤ 24 points [16,17]. Therefore, we also worked with a cutoff score of ≤ 24 points.

POBAV consists of two parts. In the first part, the patient has to name twenty pictures in writing, and in the second part, the subject has one minute to equip the memorized pictures [12]. The cognitive function domains tested are: memory, fatal functions and speed of thought. The norms for the Czech population have been determined on the basis of a valid study to date as follows: two or more errors in picture naming and five or fewer correctly equipped pictures may indicate a cognitive deficit [12]. The author of this test gave consent for its use in the research conducted.

Data collection was preceded by obtaining written consent from the management of the selected homes for the elderly to conduct the research. Data collection was carried out personally by the author of the research. Cooperation was established with the elderly home user, the nature of the research was explained and informed consent was signed with the witnesses (see ethical aspects). The standardised MoCA test was then administered, followed by a short break (approximately 10 min) and completion of the recording sheet. Finally, a second standardized test, the POBAV, was administered. The tests were performed exactly according to the prescribed standardized instructions for the test. At the end of the testing day, information (age, educational attainment, length of stay in the home, regular physical activities and cognitive training) was verified from the user's documentation and from the social worker and entered by the client in the record sheet. No uncompensated visual impairment or graphomotor impairment was noted in the research participants during the use of the POBAV test.

 

File

The research population consisted of 76 users of residential social services. Three homes for the elderly were chosen for the research. The homes were selected on the basis of several criteria: registration in the Register of Social Service Providers, consent to be included in the research, place of service provision -⁠ the capital city of Prague, provision of care for a maximum of 100 clients. The seniors from each home were deliberately selected based on availability and the given criteria. The entry criteria were: consent to participate in the research, age 65 years and over, ability to communicate verbally, no severe visual deficit.

 

Data analysis

Data obtained from the MoCA and POBAV tests were evaluated according to standardized procedures for evaluating these tests. For the MoCA test, parametric statistics tests (two-sample t-test for gender and ANOVA for education and cognitive training) were used to detect differences in the level of cognitive function depending on the categorical variables of interest (gender, education or cognitive training), as the resulting values of the MoCA test come from a normal distribution of the data. The POBAV test results (for both naming errors and correctly fitted names) come from an alternative distribution, so tests of non-parametric statistics (Mann-Whitney test for gender and Kruskal-Wallis test for education and cognitive training) were used for the above relationships. Spearman's non-linear correlation coefficient was used to calculate the correlations between the cognitive function tests (both MoCA and POBAV) and age, length of stay in the facility and fitness training, as the inter-variability of the data did not allow the application of a linear correlation coefficient calculation. In contrast, testing the correlation between the MoCA and POBAV comparisons (in both variations) fitted a linear distribution to the data. Therefore, the correlation between these tests was tested using Pearson's linear correlation coefficient. The comparison of the number of participants with cognitive norm and with cognitive deficit -⁠ within the implemented MoCA and POBAV tests -⁠ was verified through a four-field contingency table and chi-square test of independence. All of the above tests were tested at the 5% level of significance (a = 0.05).

 

Results

Characteristics of the cohort by sex, education and participation in cognitive training are presented in Table 1. Information regarding age, length of stay and participation in fitness training and MoCA test scores in the facility and POBAV are presented in Table 2. Sixteen seniors (21.1%) had normal cognitive status according to MoCA (cutoff score ≤ 24 points) (Figure 1). According to POBAV, 19 seniors (25%) had normal cognitive status (Figure 1). For MoCA and POBAV, there was no statistically significant difference in the level of cognitive function according to gender, education, or length of stay in the nursing home, but there was a statistically significant difference in the level of cognitive function according to age, cognitive training, and fitness (Table 3).

Comparing the performance of the MoCA and POBAV tests in detecting cognitive impairment/normality in institutionalized elderly, similar results were calculated. A total of 57 seniors were found to have cognitive impairment in the POBAV test, 55 of whom were also detected by the MoCA test (this concordance was confirmed in 96.5% of cases). A similar result can also be found in the case of meeting the cognitive norm, in which a total of 19 seniors were included through the POBAV test, of which 14 seniors were also included in the cognitive norm through the MoCA test (the agreement of the inclusion of seniors in the cognitive norm was confirmed in 73.7% of cases) (Table 4). This correlated distribution of responses was confirmed as significant by chi-square test (p < 0.001). At the same time, a significant correlation was also found between the MoCA and POBAV tests themselves (in both variants). The Pearson's linear correlation coefficient between the MoCA and POBAV -⁠ Picture Naming Errors tests shows a negative correlation of -0.625 (p < 0.001), indicating that the number of naming errors in the POBAV test decreases as the MoCA score increases (Figure 2). The Pearson's linear correlation coefficient between the MoCA and POBAV -⁠ Correctly Fitted Pictures tests presents a positive correlation of magnitude 0.86 (p < 0.001), indicating that as the number of MoCA scores increases, the number of correctly fitted pictures in the POBAV test also increases (Figure 3).

Based on the results of both tests from all 76 participants, a regression equation was designed to provide an estimate of the MoCA score for POBAV. In the case of the POBAV result Picture Naming Errors, the coefficient of determination is 0.391. The linear equation for this model is: MoCA = 23.59 + (-1.45) × POBAV -⁠ Picture Naming Errors. For the POBAV -⁠ Correctly fitted images results, the coefficient of determination was 0.699 and the linear equation for this model is: MoCA = 10.89 + 1.86 × POBAV -⁠ Correctly fitted images.

 

Discussion

The aim of the research was to assess cognitive function in elderly people in nursing homes using standardized tests MoCA and POBAV. A score at the level of normal cognitive status according to MoCA (with a cutoff score ≤ 24 points) was obtained in 21.1% of seniors. According to the results using the POBAV test, 25% of seniors had normal cognitive status. The findings are very similar to the results of the research by Vankova et al [18]. The authors conducted a research on a relatively large group of seniors (626 clients of homes with special regime and 351 clients of homes for the elderly) and investigated the prevalence of cognitive disorders in residential services for the elderly in the Czech Republic using the MMSE test. In residential care homes, 91% of clients were found to have cognitive impairment at the dementia depth. In a residential home for the elderly, 67% of users were found to have cognitive impairment at the depth of dementia. Vankova et al. compared the data with 2007. Compared to the original research, the number of clients with severe dementia tripled in nursing homes [18]. Kijowska and Szczerbińska investigating the prevalence of cognitive impairment in nursing home and residential home clients found that overall 65.2% of service users were identified as having cognitive impairment. This was 59.2% in care homes and 74.5% in nursing homes [19]. A study by Norwegian researchers showed that 82% of nursing home clients had dementia syndrome [20].

In our research we found a statistically significant difference in the level of cognitive functions according to MoCA and POBAV depending on age. Younger seniors show better results in the level of cognitive functions than older seniors. This result is confirmed by Luck et al. who reported older age as a risk factor for the development of FCI and AD [21]. Also, Langa et al. confirmed in their study that the risk of developing mild cognitive impairment increases with age [22].

Eid et al. reported female gender as a risk factor for developing AD [23]. Langa et al. concluded that men are at higher risk of developing AD [22]. Peteresen et al. suggest as a possible explanation that women are more likely to go directly from a normal level of cognitive function to dementia later in life, whereas men, on the other hand, show more MKP but at a younger age [24]. In our research, there was no statistically significant difference in the level of cognitive function depending on gender.

Through our research, we found that there is no difference in the level of cognitive functions according to MoCA and POBAV to the level of education. The result in our research is not in agreement with other authors [16,25,26]. The risk of developing AD is reduced in individuals with higher educational attainment [25]. Larsson et al. reported that higher levels of childhood education and lifetime educational attainment reduce the risk of dementia [26]. In a normative study by Kopecek et al. it was found that the MoCA outcome score was dependent on age and education but not on gender [16].

In their research, Harmand et al. found that the decline in cognitive function following client institutionalization was 0.7 points per year on the MMSE [27]. However, in our research, there was no statistically significant difference in the level of cognitive function according to MoCA and POBAV according to the length of time the elderly person stayed in institutional care. This implies that screening and continuous monitoring of cognitive function should be performed regularly in all clients, as the development of FMS or Alzheimer's disease can occur at any time during life in a nursing home.

Based on our research, it was found that people undergoing cognitive training and fitness exercises performed better on both tests. Seniors who engaged in more conditioning exercise performed at a higher cognitive level according to the MoCA. The POBAV test also showed the same result. Vařeková and Daďová state that in terms of positive effects on cognitive function, physical activity is recommended at a minimum of 2-3 times a week, and some authors state 4-7 times a week, with a minimum time of 20 min [28]. Ten Brinke et al. conducted a study involving 86 women with mild cognitive impairment aged 70-80 years. The experimental group receiving aerobic training twice a week had a significant increase in total hippocampal volume of 4% [29]. Angevaren et al. found in their study that some cognitive functions are more sensitive to physical activity. These were motor control, auditory attention and, to a lesser extent, information processing speed [30].

Shin et al. conducted a study on the effect of cognitive training in 40 healthy seniors and 40 seniors with FMS. The intervention group received 12 cognitive training sessions and showed improvements in attention and working memory [31]. Almeida et al. demonstrated a positive effect of cognitive training and combined training (cognitive training + physical training) on cognitive function. However, these were healthy adults and younger seniors [32]. The aim of the research by Botíková et al. was to prove the positive effect of cognitive training on cognitive function in seniors with dementia. There was an average score increase of two points in the MoCA test [33].

The results show that both tests were able to detect almost identically the threshold value in the level of cognitive functions, and therefore to warn about a possible problem. Based on these findings, a regression equation was established to estimate the resulting MoCA score to detect POBAV scores. The coefficient of determination reached better values for POBAV -⁠ Correctly Fitted Images (0.699), and therefore we would recommend the use of this model: MoCA = 10.89 + 1.86 × POBAV -⁠ Correctly fitted images.

Based on our testing, we have to confirm that the administration of the POBAV test is more user-friendly and time-saving in contrast to the MoCA test. The POBAV test requires 4-6 min to administer, but the MoCA test takes 13 ± 3 min for a healthy individual and 15 ± 3 min for a person with Alzheimer's disease [12].

Šabatová, in her report on systematic visits to homes for the elderly, states that in most of the facilities visited (homes for the elderly and homes with special regime), the state of cognitive functions was not regularly and systematically assessed and only a minimum of facilities carried out screening assessment of cognitive functions [34]. One reason for this is certainly that there are no uniform recommendations for the assessment of cognitive function in elderly people in institutional care. However, if changes in cognitive function are not detected early in the elderly, there is a relatively rapid deterioration and limitation of self-sufficiency, which in turn leads to an increased need for the care that the elderly requires. For early detection of cognitive impairment we would recommend the Czech innovative test POBAV, which has many benefits that can be appreciated in institutional care for the elderly. The POBAV is time-saving, does not require special knowledge and experience to perform the test (almost anyone can test), is freely available, tests multiple types of memory and also provides information about the psychomotor tempo and graphomotor skills of the test subject, has valid standards for the Czech senior population and is highly sensitive. However, in the case of borderline scores, it is advisable to perform a more detailed diagnostic investigation on the individual [12,35]. The MoCA tests a larger number of cognitive functions and provides a comprehensive cognitive picture. However, its use is more time-consuming, requires mandatory training and certification of the test administrator, which is paid, and must be renewed every two years [12]. Another option for testing cognitive function in clinical or social practice is electronic remote testing, which may contribute to early detection of mild cognitive impairment in older individuals and subsequent initiation of early therapy or monitoring of disease progression [36]. Polanska and Bartos proposed, developed and validated a general protocol for electronic remote testing of memory and other cognitive functions by a trained person using the ALBA test and modified POBAV and ACE-III tests via webcam and microphone. The authors report that electronic testing is comparable to in-person testing [36]. The ALBAV electronic test for self-examination of memory has also been newly developed in the Czech setting [37]. This test has already been validated and has confirmed sufficient discriminant and concurrent validity [38].

We consider the results of the presented research to be beneficial, however, the research has its limitations. We consider the main limitation of the research to be the size of the research sample; therefore, the findings should not be generalized to the entire population of seniors in residential facilities. Also, at the time of the research, the COVID-19 pandemic was ongoing and seniors in residential services missed out on many socio-cultural stimuli, group activities and therapies, and social contacts with family and friends. All these aspects have a very significant impact on cognitive function.

 

Conclusion

Early systematic screening of cognitive deficits and subsequent coordinated setting of long-term care for seniors living in residential social services represent an effective strategy that can significantly contribute to optimizing the quality of life of seniors. For early detection of cognitive deficits in institutional care for the elderly, the POBAV screening test is appropriate. As part of comprehensive care to prevent or minimise the occurrence of cognitive dysfunction, it is important to encourage seniors to engage in physical activity and motivate them to participate in cognitive training.

 

Ethical aspects

The research was carried out in accordance with the Helsinki Declaration of 1975 (and its revisions of 2004 and 2008). At the time of the research, the type of research was not subject to the internal regulations of the Ethics Committee for Research of the FSS OU (EKV). The EKV reviewed the research plan and methodology ex post and issued an approval opinion on 14 June 2023 (Ref: OU-68230/20-2023).

Participants were verbally informed about the nature of the research in an understandable way and informed consent was signed with each participant. Due to the topic, the informed consent was also signed by two witnesses each time. All data collected was handled according to applicable ethical standards and the anonymity of the research participants was maintained.

 

Conflict of interest

The authors declare that they have no conflict of interest in relation to the subject of the study.

 

Table 1. Characteristics of the cohort by gender, education and participation in cognitive training .

		Absolute frequency	Relative frequency
Gender	men	12	15,8
	Women	64	84,2
	Total	76	100,0
Education	primary school	10	13,2
	secondary vocational school	23	30,3
	high school	29	38,2
	college	14	18,4
	Total	76	100,0
participation in cognitive training	0× per week	35	46,1
	1 time per week	27	35,5
	2 times a week	12	15,8
	3 times a week	1	1,3
	≥ 4 times a week	1	1,3
	Total	76	100,0

 

 

Table 2. Characteristics of the cohort according to age, length of stay, participation in fitness training and MoCA and POBAV skin.

	Age in years	Length of stay in the facility (in months)	Conditioning exercises (weekly)	MoCA	POBAV -⁠ Picture naming errors	POBAV -⁠ Correctly fitted images
diameter	88,8	29,4	3,1	19,6	2,7	4,7
median	89,1	19,7	3,0	20,0	2,0	5,0
modus	70,5	2	2	23	1	5
standard deviation	7,5	28,6	2,4	6,0	2,6	2,7
dispersion	55,9	817,4	5,6	36,5	6,7	7,4
variation range	33,1	138	10	24	10	11
minimum	70,5	1	0	6	0	0
maximum	103,6	140	10	30	10	11

MoCA -⁠ Montreal Cognitive Assessment; POBAV -⁠ Picture Naming and Picture Aids

 

 

 Table 3. MoCA and POBAV test scores in relation to the selected variables.

	MoCA		POBAV
			Errors in the naming of images		Properly equipped pictures
	p* value	Spearman's correlation coefficient**	p* value	Spearman's correlation coefficient**	p* value	Spearman's correlation coefficient**
Age	0,00	-0,33	0,00	0,45	0,00	-0,42
fitness exercises	0,00	0,42	0,03	-0,25	0,00	0,39
cognitive training	0,00	-	0,01	-	0,00	-
Gender	0,73	-	0,58	-	0,22	-
Education	0,79	-	0,21	-	0,60	-
length of stay	0,64	0,06	0,27	0,13	0,75	0,04

*p-value is measured at the significance level α =0.05; **correlation coefficient is interpreted according to de Vause: (0.01-0.09) none; (0.10-0.29) low to moderate; (0.30-0.49) moderate to substantial; (0.50-0.69) substantial to very strong; (0.70-0.89) very strong; (0.90-0.99) almost perfect strength of association

MoCA -⁠ Montreal Cognitive Assessment; POBAV -⁠ Picture Naming and Picture Aids

 

 

Table 4. MoCA vs. POBAV skins -⁠ four-field table.

			MoCA		Total
			cognitive impairment	standard
POBAV	cognitive impairment	absolute frequency	55	2	57
		% of POBAV	96,5 %	3,5 %	100,0 %
		% of all	72,4 %	2,6 %	75,0 %
	standard	absolute frequency	5	14	19
		% of POBAV	26,3 %	73,7 %	100,0 %
		% of all	6,6 %	18,4 %	25,0 %
Total		absolute frequency	60	16	76
		% of POBAV	78,9 %	21,1 %	100,0 %
		% of all	78,9 %	21,1 %	100,0 %

MoCA -⁠ Montreal Cognitive Assessment; POBAV -⁠ Picture Naming and Picture Aids